The purpose of this report is to describe the course, Analysis of the Real Line, taught at The University of Texas at Austin. Course materials are presented using the inquiry based learning method. Students work a series of warm up problems before being presented rigorous problems in calculus, including topics on integration, exponential functions, and real number line analysis. Additionally, students consider aspects of these problems that could be incorporated into a high school curriculum. Typical problems in several major areas are summarized along with warm up problems that introduce or extend the topics. / text

Η παρούσα διατριβή εντάσσεται ερευνητικά στην περιοχή της Στατιστικής Θεωρίας Αποφάσεων και ειδικότερα στην (σημειακή) εκτίμηση του ποσοστιαίου σημείου στο μοντέλο της διπαραμετρικής εκθετικής κατανομής. Το πρόβλημα της εκτίμησης του ποσοστιαίου σημείου από τη σκοπιά της Στατιστικής Θεωρίας Αποφάσεων ακολούθησε αυτό της παραμέτρου κλίμακας, ειδικότερα αναφέρουμε το πρόβλημα εκτίμησης της διασποράς κανονικής κατανομής με άγνωστη μέση τιμή από τον Stein (1964). Στην εργασία εκείνη ο Stein απέδειξε ότι, με κριτήριο το μέσο τετραγωνικό σφάλμα, ο βέλτιστος αναλλοίωτος εκτιμητής της διασποράς είναι μη αποδεκτός, κατασκευάζοντας άλλον με μικρότερο μέσο τετραγωνικό σφάλμα. Εν συνεχεία, οι Brewster and Zidek (1974) παρουσίασαν δύο γενικές τεχνικές κατασκευής βελτιωμένων εκτιμητών, εφαρμόσιμες για τυχαία bowl-shaped συνάρτηση ζημίας και αποτελεσματικές, κυρίως όταν η υπό εκτίμηση παράμετρος είναι η παράμετρος κλίμακας και επί πλέον υπάρχει και άλλη άγνωστη παράμετρος. Αντικείμενο της μεταπτυχιακής διατριβής είναι η εκτίμηση του ποσοστιαίου σημείου θεωρώντας ανεξάρτητα τυχαία δείγματα από εκθετικούς πληθυσμούς με την ίδια παράμετρο θέσης και διαφορετική παράμετρο κλίμακας για κάθε πληθυσμό ξεχωριστά. Βασιζόμενοι στην εργασία των Kumar and Sharma (1996) βρίσκουμε εκτιμητή μέγιστης πιθανοφάνειας και αμερόληπτο εκτιμητή ελάχιστης διασποράς για το ποσοστιαίο σημείο από τον πρώτο εκθετικό πληθυσμό και στην συνέχεια εφαρμόζουμε τη τεχνική κατασκευής, βελτιωμένων εκτιμητών, των Brewster and Zidek (1974). Η παρουσίαση των επί μέρους θεμάτων και αποτελεσμάτων της διατριβής αυτής οργανώνεται ως εξής. Στο Κεφάλαιο 1 αναφέρονται κάποια βασικά στοιχεία θεωρίας από τη Μαθηματική Στατιστική, όπως βασικοί ορισμοί και θεωρήματα σχετικά κυρίως με τη συνάρτηση κινδύνου (risk function), τους εκτιμητές (UMVUE), τους εκτιμητές μέγιστης πιθανοφάνειας (MLE) και τους αναλλοίωτους (equivariant) εκτιμητές. Στο Κεφάλαιο 2 ορίζεται η διπαραμετρική εκθετική κατανομή και το ποσοστιαίο σημείο της διπαραμετρικής εκθετική κατανομής, , θετική σταθερά ,από τον πρώτο εκθετικό πληθυσμό, το οποίο στη συνέχεια εκτιμάται από τον εκτιμητή μέγιστης πιθανοφάνειας και από τον εκτιμητή. Στο Κεφάλαιο 3 χρησιμοποιούνται τεχνικές βελτίωσης του εκτιμητή του ποσοστιαίου σημείου. Αρχικά εντοπίζεται ο βέλτιστος εκτιμητής του ποσοστιαίου σημείου στην κλάση των εκτιμητών με κριτήριο το μέσο τετραγωνικό σφάλμα και στη συνέχεια χρησιμοποιείται η τεχνική κατασκευής, βελτιωμένων εκτιμητών, των Brewster and Zidek (1974) όταν και όταν . Τέλος στο Κεφάλαιο 4 αναφέρονται κάποια Λήμματα τα οποία χρησιμοποιούνται σε αποδείξεις προτάσεων της διατριβής. / Estimating quantiles of a selected exponential population from k populations.

No description available.

This paper describes various network monitoring technologies and anomaly detection methods. NetFlow were chosen for anomaly detection system being developed. Anomalies are detected using a deviation value. After evaluating quality of developed system, new enhancements were suggested and implemented. Flow data distribution was suggested, to achieve more precise NetFlow data representation, enabling a more precise network monitoring information usage for anomaly detection. Arithmetic average calculations were replaced with more flexible Exponential Weighted Moving Average algorithm. Deviation weight was introduced to reduce false alarms. Results from experiment with real life data showed that proposed changes increased precision of NetFlow based anomaly detection system.

The ring of polynomials over a finite field has many arithmetic properties similar to those of the ring of rational integers. In this thesis, we apply the Hardy-Littlewood circle method to investigate the density of rational points on certain algebraic varieties in function fields. The aim is to establish asymptotic relations that are relatively robust to changes in the characteristic of the base finite field. More notably, in the case when the characteristic is "small", the results are sharper than their integer analogues.

In 1974, H. Ruderman posed the following question: If $(2^m-2^n)|(3^m-3^n)$, then does it follow that $(2^m-2^n)|(x^m-x^n)$ for every integer $x$? This problem is still open. However, in 2011, M. R. Murty and V. K. Murty showed that there are only finitely many $(m,n)$ for which the hypothesis holds. In this thesis, we examine two generalizations of this problem. The first is replacing 2 and 3 with arbitrary integers $a$ and $b$. The second is to replace 2 and 3 with arbitrary algebraic integers from an imaginary quadratic field. In both of these cases we have shown that there are only finitely many $(m,n)$ for which the hypothesis holds. To get the second result we also generalized a result by Bugeaud, Corvaja and Zannier from the integers to imaginary quadratic fields. In the last half of the thesis we use the abc conjecture and some related conjectures to study some exponential Diophantine equations. We study the Pillai conjecture and the Erd\"{o}s-Woods conjecture and show that they are implied by the abc conjecture and that when we use an effective version, very clean bounds for the conjectures are implied. / Thesis (Ph.D, Mathematics & Statistics) -- Queen's University, 2012-04-13 12:04:14.252

Many production systems have split and merge operations to increase production capac- ity and variety, improve product quality, and implement product control and scheduling policies. This thesis presents analytical methods to model and analyze split and merge production systems with Bernoulli and exponential reliability machines under circulate, priority and percentage policies. The recursive procedures for performance analysis are de- rived, and the convergence of the procedures and uniqueness of the solutions, along with the structural properties, are proved analytically, and the accuracy of the estimation is justi¯ed numerically with high precision. In addition, comparisons among the e®ects of di®erent policies in system performance are carried out.

An open queueing network model (QNM) is proposed for wormhole-routed hypercubes with finite buffers and deterministic routing subject to a compound Poisson arrival process (CPP) with geometrically distributed batches or, equivalently, a generalised exponential (GE) interarrival time distribution. The GE/G/1/K queue and appropriate GE-type flow formulae are adopted, as cost-effective building blocks, in a queue-by-queue decomposition of the entire network. Consequently, analytic expressions for the channel holding time, buffering delay, contention blocking and mean message latency are determined. The validity of the analytic approximations is demonstrated against results obtained through simulation experiments. Moreover, it is shown that the wormholerouted hypercubes suffer progressive performance degradation with increasing traffic variability (burstiness).

Priority based scheduling disciplines are widely used by existing computer operating systems. However, the mathematical analysis and modelling of these systems present great difficulties since priority schedulling is not compatible with exact product form solutions of queueing network models (QNM's). It is therefore, necessary to employ credible approximate techniques for solving QNM's with priority classes. The principle of maximum entropy (ME) is a method of inference for estimating a probability distribution given prior information in the form of expected values. This principle is applied, based on marginal utilisation, mean queue length and idle state probability constraints, to characterise new product-form approximations for general open and closed QNM's with priority (preemptive-resume, non-preemtive head-of-line) and non-priority (first-come-first-served, processor-sharing, last-come-first-served with, or without preemtion) servers. The ME solutions are interpreted in terms of a decomposition of the original network into individual stable GIG11 queueing stations with assumed renewal arrival processes. These solutions are implemented by making use of the generalised exponential (GE) distributional model to approximate the interarrival-time and service-time distributions in the network. As a consequence the ME queue length distribution of the stable GE/GEzl priority queue, subject to mean value constraints obtained via classical queueing theory on bulk queues, is used as a 'building block' together with corresponding universal approximate flow formulae for the analysis of general QNM's with priorities. The credibility of the ME method is demonstrated with illustrative numerical examples and favourable comparisons against exact, simulation and other approximate methods are made.

Economic stringencies have compelled Canadian hospitals to examine their administrative effectiveness critically. Improved supplies and inventory procedures adopted by leading industrial corporations, suggest that hospitals might benefit from such systems. Lack of the profit incentive, and the high ratio of wages to total expenses in hospitals, have delayed adoption of modern inventory management techniques. This study examined the economic status of Canadian hospitals, and endeavoured to discover whether a computer-based inventory management system, incorporating short-term statistical demand forecasting, would be feasible and advantageous. Scientific forecasting for inventory management is not used by hospitals. The writer considered which technique would be most suited to their needs, taking account of benefits claimed by industrial users. Samples of demand data were subjected to a variety of simple forecasting methods, including moving averages, exponentially smoothed averages and the Box-Jenkins method. Comparisons were made in terms of relative size of forecast errors; ease of data maintenance, and demands upon hospital clerical staffs. The computer system: BRUFICH facilitated scrutiny of the effect of each technique upon major components of the system. It is concluded that either of two methods would be appropriate: moving averages and double exponential smoothing. The latter, when combined with adaptive control through tracking signals, is easily incorporated within the total inventory system. It requires only a short run of data, tracks trend satisfactorily, and demands little operator intervention. The original system designed by this writer was adopted by the Hospital for Sick Children, Toronto, and has significantly improved their inventory management.